cs.CL（2026-04-22）

📊 共 19 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Intersectional Fairness in Large Language Models 系统性评估大型语言模型在交叉人口属性下的公平性问题 large language model
2 The GaoYao Benchmark: A Comprehensive Framework for Evaluating Multilingual and Multicultural Abilities of Large Language Models GaoYao：构建多语言文化能力评测基准，诊断大语言模型全球适用性 large language model ✅
3 Less Languages, Less Tokens: An Efficient Unified Logic Cross-lingual Chain-of-Thought Reasoning Framework 提出UL-XCoT框架，通过减少语言和token数量提升跨语言CoT推理效率。 chain-of-thought
4 DialToM: A Theory of Mind Benchmark for Forecasting State-Driven Dialogue Trajectories 提出DialToM基准测试，用于评估LLM在对话轨迹预测中的心智理论能力。 large language model ✅
5 Toward Cross-Lingual Quality Classifiers for Multilingual Pretraining Data Selection 提出跨语言质量分类器，用于多语言预训练数据选择，提升低资源语言模型质量。 large language model
6 Enhancing Research Idea Generation through Combinatorial Innovation and Multi-Agent Iterative Search Strategies 提出基于组合创新和多智能体迭代搜索的研究想法生成框架，提升想法多样性和新颖性 large language model ✅
7 Knowledge Capsules: Structured Nonparametric Memory Units for LLMs 提出知识胶囊，通过外部键值注入增强LLM在长文本和多跳推理中的知识利用。 large language model
8 Cooperative Profiles Predict Multi-Agent LLM Team Performance in AI for Science Workflows 提出合作特征以预测多智能体LLM团队在科学工作流中的表现 large language model
9 HaS: Accelerating RAG through Homology-Aware Speculative Retrieval 提出HaS框架以加速RAG检索过程 large language model ✅
10 All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG 提出LAURA以解决多语言RAG中存在的语言偏见问题，提升跨语言检索增强生成性能。 large language model
11 Dual-Cluster Memory Agent: Resolving Multi-Paradigm Ambiguity in Optimization Problem Solving 提出双簇记忆代理（DCM-Agent），解决优化问题中多范式歧义性难题。 large language model
12 Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives 提出基于LLM摘要的偏见分析流程，用于评估LLM在生命叙事解读中的种族和性别偏见。 large language model
13 To Know is to Construct: Schema-Constrained Generation for Agent Memory 提出SCG-MEM，通过模式约束生成实现Agent Memory的有效访问，解决结构性幻觉问题。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
14 WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning WebGen-R1：利用强化学习激励大语言模型生成功能完备且美观的网站 reinforcement learning large language model multimodal
15 SAKE: Self-aware Knowledge Exploitation-Exploration for Grounded Multimodal Named Entity Recognition SAKE：面向GMNER的自感知知识探索与利用框架，解决长尾实体识别难题。 reinforcement learning large language model multimodal
16 Multi-Perspective Evidence Synthesis and Reasoning for Unsupervised Multimodal Entity Linking 提出MSR-MEL框架，利用多视角证据合成与推理，解决无监督多模态实体链接问题。 teacher-student large language model multimodal
17 AFMRL: Attribute-Enhanced Fine-Grained Multi-Modal Representation Learning in E-commerce 提出AFMRL，通过属性增强细粒度多模态表征学习解决电商场景下的商品检索难题。 representation learning contrastive learning large language model
18 LayerTracer: A Joint Task-Particle and Vulnerable-Layer Analysis framework for Arbitrary Large Language Model Architectures LayerTracer：面向任意LLM架构的任务知识定位与脆弱层分析框架 Mamba large language model
19 Hybrid Policy Distillation for LLMs 提出混合策略蒸馏（HPD）方法，提升LLM在长文本生成和代码任务上的性能。 distillation large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页